Recycle absorber

Discusses use of COCO, the process simulation and modelling software suite from AmsterCHEM, downloadable from http://www.cocosimulator.org

Moderator: jasper

Recycle absorber

Postby kevinmil » 04 April 2016, 14:14

Hello,
I would like to congratulate all of you for this incredible simulator.
I would like to simulate an absorber with make up and recycle loop. The results differ a lot when you change the recycle tolerance in the settings. Is this normal?
On the other hand, I want to build a controller in order to fix stream 75 by varying the flow in steam 203 but I do not know how to connect the output in the controller since I cannot do this.

I also suggest that it should be really useful to be able to specify the splitter with the flow of one of the stream in case that the splitter has 2 streams, not just with the split fraction.

Thank you very much in advance.
Attachments
Doubt COCO.fsd
(71.01 KiB) Downloaded 743 times
kevinmil
 
Posts: 2
Joined: 14 March 2016, 11:18

Re: Recycle absorber

Postby jasper » 04 April 2016, 16:28

A recycle tolerance of 4e-3 is rather high, but still, if you evaluate all units in the flowsheet, starting at the cut stream, you should - at solution - not find any component flow, ln(T) or ln(P) (these are COFE's internal variables for a stream while solving) that change by a factor more than 1 plus or minus 4e-3 when evaluating all units in the recycle.

This does of course not imply that you are within a relative error of 4e-3 from the solution. The solver cannot determine this as it does not know the solution. As with most iterative solver the error is specified on the function evaluation, not on the solution.

Here's how to control stream 75, presuming you want to have it below the flow rate of stream 7:

- delete streams 201(2), 203(2) and the make-up mixer
- place a splitter, connect 7(2) as feed, and 8(2) as outlet 1
- create another stream to connect as outlet 2
- disconnect any of the streams between the start of 8(2) and 75(2). As these all have the same mass flow, it makes sense to disconnect close to the mixer to make the calculation recycle small. E.g. disconnect 8(2) from the pump.
- insert a measure unit, connect 8(2) as the feed, connect the product to the pump with a new stream.
- edit the measure unit, on the Measure tab, hit Add, add Total Mass Flow (or total mole flow, depending on your interest)
- insert a controller
- use an information stream to connect the measured mass flow to the controller
- double click the mixer. On the Ports tab, Port Operations, Add Virtual Information Port. Select Input, select Split Factors. As this is an array, you must specify the index of desire. Accept the default 0.
- connect the contoller variable of the controller to the new split factor info port of the splitter using an information stream
- double click this stream, set the initial guess of the split factor
- double click the controller to set the flow rate set point
- hit solve

It is not straight forward to specify something that makes sense; the system easily diverges to very high flow rates.
User avatar
jasper
 
Posts: 1128
Joined: 24 October 2012, 15:33
Location: Spain

Re: Recycle absorber

Postby kevinmil » 05 April 2016, 13:05

Hi Jasper,

Thank you for your quick answer. However, I continue stuck because I cannot fulfill all your steps. But I have three doubts:
1) I have already removed the makeup mixer. So when you say mixer I do not know where it is placed.
2) I do not know how to connect the output information stream of the controller. The system does not allow me no connect it.
3) Do you mean flow constraint or measure unit? According to the options that you mentioned me it seems that you mean flow constraint.

Could you send me a file with an example of a controller?

Thank you very much in advance.
Best regards,
Kevin
kevinmil
 
Posts: 2
Joined: 14 March 2016, 11:18

Re: Recycle absorber

Postby jasper » 05 April 2016, 16:26

1) I am not clear on this question - elaborate?

2) The controller should always allow for connecting information streams, one in and one out. If it does not connect, the reason should show in the log window when you are trying to connect.

3) a measure unit. The flow constraint is not required - the system is closed by manipulating the split ratio. I am also not sure how this could be set up with a flow constraint, as this would impose the flow rate after the device, and there is no way for it to force the splitter to do the right thing. A flow constraint is intended for 1) controlling a feed rate (so not inside a recycle) or 2) controlling the flow rate in a closed loop (without feeds or products)

Here's a controller example: http://cocosimulator.org/down.php?dl=controller.fsd

Here's an example of how to use the flow constraint to control feed rate: http://cocosimulator.org/down.php?dl=feed_controller.fsd
User avatar
jasper
 
Posts: 1128
Joined: 24 October 2012, 15:33
Location: Spain

Re: Recycle absorber

Postby Usseglio » 18 November 2017, 22:23

Hello,

Sorry, it´s been a while since these posts, but I always had this configuration in mind.
I have set up a similar configuration also for an absorber I used for different type of process simulations, and it is working quite well.

Kevin, I downloaded your original file but I could not open it properly from current COCO version.
Have you already solved your configuration ?
If not, if you want upload the file again to share with you guys the approach I have used for this process configuration.

Kind regards,

Marcelo
Usseglio
 
Posts: 34
Joined: 23 November 2012, 14:16


Return to COCO (AmsterCHEM)

Who is online

Users browsing this forum: No registered users and 6 guests

cron